ace19-dev / gvcnn-tf Goto Github PK

Excuse me, why does “validation_batch_xs” in train.py has only one decimal place?

I am wondering the highest accuracy of your code on ModelNet 40. Could your share the detailed information?

hello,thanks for your code.There is a question about filter = ['1','2','4','7','8','10'] in the train.py,i don't know what's that.Thank you for your reply.

Would you please give a more detail explanation for data preparation and train/test configurations?
Thanks a lot!

I want to use the retrieval method of this article, you seem to implement it, but there is no code in the retrieval file.

Hello,
I was trying to validate the dataset with pre-trained checkpoints. But it seems like there is missing function in nets/model class which is being accessed by eval.py class.

I got the following error.
"AttributeError: module 'nets.model' has no attribute 'discrimination_score".

Is there something I am missing or else?

Thank you!

Hi,

i just tried to train the gvcnn model using the standard steps you expained. I downloaded the modelnet10 data, used data_utils for .png creation and dataset_tools for the creation of tfrecord.

For training I used the standard settings from train.py but also after two days of training there is no learning progress at all. Is there anything special to consider?

Best regards,
Tobi

Could you achieve the results of the paper?

After preparing for the modelnet40, there is an error when running the code:

totalMemory: 10.91GiB freeMemory: 10.34GiB
2019-04-22 10:37:55.656508: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1511] Adding visible gpu devices: 0
2019-04-22 10:37:55.782679: I tensorflow/core/common_runtime/gpu/gpu_device.cc:982] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-04-22 10:37:55.782723: I tensorflow/core/common_runtime/gpu/gpu_device.cc:988] 0
2019-04-22 10:37:55.782730: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 0: N
2019-04-22 10:37:55.782912: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 9995 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1080 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1)

Epoch 0

2019-04-22 10:38:30.294264: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.299554: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.308661: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.314694: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.319719: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.332460: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.338507: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.346839: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.350631: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.366402: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.370935: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.378067: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.386358: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.392612: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.397588: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.407753: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.416823: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.430087: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.439906: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.450038: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.454743: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
2019-04-22 10:38:30.460643: W tensorflow/core/framework/op_kernel.cc:1273] OP_REQUIRES failed at example_parsing_ops.cc:240 : Invalid argument: Key: image/encoded. Can't parse serialized Example.
Traceback (most recent call last):
File "/home/arcwoc/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1334, in _do_call
return fn(*args)
File "/home/arcwoc/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1319, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "/home/arcwoc/anaconda3/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1407, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Key: image/encoded. Can't parse serialized Example.
[[{{node ParseSingleExample/ParseSingleExample}} = ParseSingleExample[Tdense=[DT_STRING, DT_INT64], dense_keys=["image/encoded", "image/label"], dense_shapes=[[8], []], num_sparse=0, sparse_keys=[], sparse_types=[]](arg0, ParseSingleExample/Const, ParseSingleExample/Const_1)]]
[[{{node IteratorGetNext}} = IteratorGetNextoutput_shapes=[[?,8,299,299,3], [?]], output_types=[DT_FLOAT, DT_INT64], _device="/job:localhost/replica:0/task:0/device:CPU:0"]]

I have followed the steps in the ReadMe, but i do not know how this happens and i am looking forward for your solution, thanks a lot!

Hi,

May I ask how the grouping module is trained? Is it trained as part of the end-to-end model or it is trained separately?

Thanks!